This short note is a preliminary one, which contains the observational results obtained during last summer in Tokyo by means of a simple recorder of Simpson's type (Pron. Roy. Soc. A 161 309-352, 1937). The antenna used is about 70m long, one end being about 30m high, the other 69m. Three near thunderstorms were investigated which were weak and originated without exception in the vicinity of Tokyo and had almost negative atmospheric potential gradients on the ground. Three other remote thunderstorms were also recorded, potential gradients being more often negative than positive. So far as the present results are concerned. the electric fields under thunderclouds are completely explained by Simpson's model. It is also noteworthy that several hours before the attack of a typhoon in Tõkyõ, a negative electric field was recorded. It was already noticed among Japanese forecasters that a typhoon was often preceeded by thunderstorms. The above record is thus interesting in connection with this fact together with Arakawa's hypothesis of typhoon rainfall, which, after him, must be accomplished by some instability mechanism.
